Barrett?s esophagus (BE) is the dangerous obligate premalignant precursor lesion of esophageal adenocarcinoma (EAC), one of the most rapidly increasing and lethal malignancies in the United States and Europe(3). EAC is rarely detected before it becomes invasive and untreatable, and 95% of EACs develop in patients not previously diagnosed with BE(19)(20). Patients diagnosed with BE, in contrast, have an excellent prognosis, since neoplasia is detected very early by periodic endoscopic surveillance. There is, however, no currently available screening test for BE. Clinical translation of minimally invasive, low-cost biomarker-based approaches to BE diagnosis will improve early EAC detection and increase overall survival. By combining our swallowable, retrievable esophageal sample collection sponge (EsophaCapTM) manufactured by our industrial partner, Capnostics, with our patented BE DNA methylation markers and our enhanced processing technique, Methylation-On-Beads (MOB) to maximize extraction and bisulfite conversion of DNA; and by establishing this assay in the CLIA-compliant laboratory of our industrial partner, MyGenetx, we can now make this assay widely available. Preliminary data demonstrate high diagnostic accuracy of our sponge-based biomarker test for BE. We will apply this strategy by pursuing the following Specific Aims: Aim 1. To analytically validate our EsophaCapTM-based BE diagnostic assay. Aim 1a. First, using technical replicate EsophaCapTM specimens from 42 newly recruited BE patients and 42 non-BE controls, we will confirm the accuracy, robustness, and reliability of our BE diagnostic assay. Preliminary results in this context are also encouraging (see Table 2, Technical replicates). Aim 1b. Next, in these same 42 patients with known BE vs. 42 controls without BE, we will establish analytical concordance of EsophaCapTM-based data with matched tissue biopsy sample data. Aim 2. To conduct a pilot study to validate a multivariate model for EsophaCapTM-based diagnosis of BE. Based on our already-collected EsophaCapTM-derived specimen training dataset (see Fig. 5), we have constructed a 3-marker prediction model for BE (Fig. 6). We will apply this model and the chosen cut- off threshold to a newly collected set of 50 untested samples (independent of the Aim 1 samples). Aim 3. To prospectively test the combined sponge-methylation biomarker strategy in a test set cohort of BE patients vs. controls. Our assay will be performed in EsophaCapTM-derived samples from a prospectively- collected cohort of 80 BE patients and 240 controls. The multivariate model validated in Aim 2 will be applied to this test set of large patient population. Aim 4. To industrialize our EsophaCapTM assay protocol. In parallel and simultaneously with Aims 1-3, we will establish all steps in our MOB-based DNA extraction, bisulfite modification and quantitative methylation-specific PCR (qMSP) protocol in the CLIA-compliant laboratory at MyGenetx. Assays performed in Aim 3 will be repeated by obtaining repeat sponges from the same patients, this time in the CLIA lab, and checked for accuracy by comparison to Aim 3 results.